Statistical Meanderings, 2024
Below I share some observations from reviewing the “End of Season Statistics” I published in October. They are not particularly analytical in nature, just oddities and curiosities that I noted when casually perusing the reports. Almost all figures quoted are from said reports and as a reminder they include park adjustment which is why they might vary from what you see elsewhere in the case of something like a slash line.
* The team with the highest run scoring environment during its games was not Colorado, the usual leader (they led in each of the last two seasons), but rather Arizona. The D-Backs and their opponents average 10.10 runs per game in the first nine innings, with the Rockies coming in second at 9.71. Arizona did this by leading the majors in runs scored in regulation by a wide margin while finishing fifth in runs allowed. Interestingly, the park factor for Chase Field has drifted down to 1.01 using 2020-2024; historically it has often been around 1.05 which would help inflate the run environment even more.
* While the World Series provided one absolute classic and one epic meltdown, it was a bit disappointing that the Dodgers and Yankees didn’t produce an overall classic series. One can quibble about best record v. best team, but I would say it was the first time since 2020 that the two best teams in each league met, and that 2018, 2017, and 2013 were the last times in a full season that one could make that case. LA was #1 in W%, NYA 3rd (PHI 2nd); NYA first in EW%, LA third (MIL snuck in at second); and LA first, NYA 2nd in PW%, so however you slice it there is a strong case to be made that at least over the sample of the regular season, they were the two best.
They also had the profile of (classically) sabermetric-friendly offenses, tying for the lead in Secondary Average and finishing 2/3 in walk rate (SEA first, NYA second) and 1/3 in ISO (LA first, BAL second).
* Cleveland and Milwaukee finished 28th and 29th in MLB in Quality Start %. The Guardians had the fifth-highest starting pitcher eRA, while the Brewers were closer to average at eleventh. Neither team got much length, finishing seventh and fifth lowest in IP/S. This traditionally has not been the formula for winning division titles as both teams did, and is also interesting because the Brewers had the best starter eRA in 2023 and were seventh in IP/S. But the teams finished one-two in reliever eRA (Cleveland nearly a half run better than Milwaukee). Guardians relievers ranked first, third, fifth, and seventh in the AL in RAR.
* The White Sox underlying performance was not as bad as their record (.253 W%, .298 EW%, .314 PW%), which is not a surprise as to set the record for most losses you would expect a terrible team that also had some “bad luck”. I’m not sure it was widely recognized (although perhaps not as poorly-educated Chris Getz has alluded to it in discussing the haul they might seek for Garret Crochet) that it was their offense that truly made them atrocious. The Sox allowed 4.86 runs per game (park adjusted but ignoring extra innings), which was third-worst in the majors but only .17 worse than the contending D-Backs (yes, they had the best offense in MLB which is why they were a contender). But the Sox 3.04 R/G was .64 lower than the next worse offense in Colorado (who were dead last in RA/G).
Chicago hit .219/.278/.338 as a team, last in BA/OBA/SLG, walk rate, and ISO. Among position players with 250 PA, the closest match to that level of futility was Jonny DeLuca (.220/.282/.336).
* I’ve already indirectly referenced Arizona’s shockingly good offense twice, but it’s worth reflecting on it in more depth. The D-Backs 5.30 R/G was first by a healthy margin over LA (5.12), and of their 11 hitters with 250 or more PA, only one was below league average in RG (Kevin Newman, -3 RAA in 303 PA). Ketel Marte was third in the NL in RG and second among qualifiers, trailing only Shohei Ohtani.
* Due to the way I classify relief pitchers (pitchers who have fewer than 15 starts and have at least 40 appearances or pitched 50 or more innings with more than half of their appearances in relief) and the fact that I don’t give any credit for leverage, sometimes the relief RAR leaderboard has some unexpected names. That was the case for the NL this season, as Nick Martinez is tops the list with 27 RAR. Ryan Yarbrough and Luis Oritz also appear in the top ten, alongside more traditionally used relievers (although Ryan Walker, Bryan Hudson, and Jose Butto are unexpected entries in that vein).
* Being a starting pitcher can be an unforgiving business. One AL starter was exactly at replacement level this season with 0 RAR – Justin Verlander. The list of NL starters below replacement level was Patrick Corbin, Walker Buehler, Kyle Hendricks, Jordan Montgomery, Roddery Munoz, and Taijuan Walker.
* This falls under the category of “oddity signifying nothing”, but the top nine AL hitters on my RAR list are also the only nine who were credited with 100 or more runs created.
* The National League had a paucity of hitters performing at a high level; while six AL hitters with 600 or more PA had a RG over 7, only one NL hitter did (Ohtani). Dropping the PA threshold down to 300 allows Christian Yelich to join the list two; alternatively, lowering the bar Ketel Marte and Marcell Ozuna were both within .11 RG. As a result, Ohtani actually led his fellow hitters in RAR by a wider margin that Aaron Judge despite Judge significantly exceeding his total; Ohtani led by 37 over Franciso Lindor (95 to 58), while Judge’s lead was 31 (109 to 78 over outfield neighbor Juan Soto).
* I continue to find the Pablo Lopez/Luis Arraez fascinating. Lopez is a pitcher whose peripherals often outpace his outcomes, and there is always the temptation to think that he will get even better and win a Cy Young award. Arraez is the posterchild for batting average continue to be overrated, providing little secondary offense at left side of the defensive spectrum positions without providing much as a fielder or baserunner.
This year, I have Lopez at 24 RAR, 28 eRAR, and 38 dRAR (outpitching the peripherals yet again). Arrarez had 22 RAR, with zero baserunning runs per Statcast. His fielding metrics were not good (-6 DRS, -5 UZR, and -9 Statcast RAA), which for this purpose I’ll average and halve for -3 runs. That brings Arraez’ total to 19 RAR. Last year Lopez led Arraez 47 to 40, so the value gap remained similar but both contributed less to their teams.